Dockerfile

# Use Python 3.10.14 slim for reproducibility and compatibility
FROM python:3.10.14-slim

# Set working directory early to avoid repeated layer creation
WORKDIR /app

# Update package sources to include contrib and non-free repositories
RUN echo "deb http://deb.debian.org/debian bookworm main contrib non-free" > /etc/apt/sources.list && \
    echo "deb http://deb.debian.org/debian-security bookworm-security main contrib non-free" >> /etc/apt/sources.list && \
    echo "deb http://deb.debian.org/debian bookworm-updates main contrib non-free" >> /etc/apt/sources.list

# Install system dependencies in a single layer to reduce size
RUN apt-get update && apt-get install -y --no-install-recommends \
    tesseract-ocr=5.3.* \
    tesseract-ocr-ben \
    poppler-utils \
    libpoppler-dev \
    curl \
    chromium \
    chromium-driver \
    libjpeg-dev \
    zlib1g-dev \
    libpng-dev \
    gcc \
    python3-dev \
    libpython3-dev \
    g++ \
    procps \
    && apt-get clean \
    && rm -rf /var/lib/apt/lists/*

# Upgrade pip to latest version
RUN pip install --no-cache-dir --upgrade pip==25.1.1

# Ensure tessdata directory and download ben.traineddata if missing
RUN mkdir -p /usr/share/tesseract-ocr/5/tessdata && \
    if [ ! -f /usr/share/tesseract-ocr/5/tessdata/ben.traineddata ]; then \
        curl -L -o /usr/share/tesseract-ocr/5/tessdata/ben.traineddata \
        https://github.com/tesseract-ocr/tessdata_best/raw/main/ben.traineddata; \
    fi

# Set TESSDATA_PREFIX
ENV TESSDATA_PREFIX=/usr/share/tesseract-ocr/5/tessdata

# Ensure chromedriver is available
RUN ln -sf /usr/lib/chromium/chromedriver /usr/bin/chromedriver || \
    ( [ -f /usr/bin/chromedriver ] && echo "chromedriver found" ) || \
    ( echo "chromedriver not found" && exit 1 )

# Set environment variables for Chromium and chromedriver
ENV CHROMIUM_PATH=/usr/bin/chromium \
    CHROMEDRIVER_PATH=/usr/bin/chromedriver \
    SELENIUM_HEADLESS=1

# Set permissions for non-root user
RUN chmod -R 777 /app && chown -R nobody:nogroup /app

# Copy and install requirements
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt && \
    find /usr/local/lib/python3.10 -name "*.pyc" -delete

# Set Hugging Face cache and create model directory
ENV HF_HOME=/data/models \
    HF_HUB_CACHE=/data/models
RUN mkdir -p /data/models && chmod -R 777 /data/models

# Pre-download the model with retries
RUN for attempt in 1 2 3; do \
    echo "Attempt $attempt to download model..." && \
    python -c "import os; os.environ['TRANSFORMERS_VERBOSITY']='error'; \
    from transformers import AutoTokenizer, AutoModelForSeq2SeqLM; \
    AutoTokenizer.from_pretrained('Helsinki-NLP/opus-mt-bn-en', cache_dir='/data/models'); \
    AutoModelForSeq2SeqLM.from_pretrained('Helsinki-NLP/opus-mt-bn-en', cache_dir='/data/models');" && \
    break || sleep 10; \
done || ( echo "Model download failed after 3 attempts" && exit 1 )

# Copy application files
COPY app.py init_db.py db_utils.py ./
COPY templates/ templates/
COPY static/ static/

# Create cache directory for OCR
RUN mkdir -p /tmp/ocr_cache && chmod -R 777 /tmp/ocr_cache

# Expose port
EXPOSE 7860

# Run with gunicorn, single worker, 900-second timeout
CMD ["gunicorn", "--workers", "1", "--bind", "0.0.0.0:7860", "--preload", "--timeout", "900", "--log-level", "info", "app:app"]